1 Yunnan Transportation Research Institute Co., Ltd., Kunming 650011, China
2 Faculty of Transportation Engineering, Kunming University of Science and Technology, Kunming 650500, China
Show less
文章历史+
Received
Published
Revised
2025-10-10
2026-08-28
2025-12-16
Online Date
2026-09-04
PDF (11982K)
摘要
隧道结构长期服役过程中易出现渗水、裂缝、剥落等表观病害,严重威胁行车安全与结构耐久性。现有隧道病害检测方法在多尺度特征提取与嵌入式部署适应性方面仍存在不足。针对这一问题,提出了一种改进型YOLOv11-WTseg隧道病害检测方法。该方法在 YOLOv11-seg 算法的解析分支中引入小波变换卷积模块(wavelet transform convolution block, WTConv),扩大模型感受野并增强多尺度特征捕获能力;同时,将边界框回归损失函数由完全交并比(complete intersection over union,CIoU)替换为SIoU(Scylla intersection over union),通过角度惩罚机制优化边界框回归过程。基于自建隧道病害数据集,选取侧沟损伤、非侧沟板损伤、路面裂缝、渗漏水和拱顶剥落作为研究对象开展对比实验。结果表明:在引入WTConv后Box_mAP从0.821提升至0.844,Mask_mAP从0.787提升至0.820;进一步采用SIoU后,Box_mAP 提升至0.845,Mask_mAP提升至0.839。与基线模型相比,改进模型在不同输入尺度下均表现出更优的检测与分割精度。同时,在NVIDIA Jetson Xavier NX平台上进行的嵌入式部署验证显示,改进模型在保持模型规模仅为5.4 M的条件下,推理延迟为117 ms,兼具轻量化特性与实时性。研究结果表明,YOLOv11-WTseg模型能够实现隧道典型病害的高效、精准检测与分割,并具备良好的工程应用前景,可为隧道结构健康监测与养护决策提供技术支撑。
Abstract
Tunnel structures are prone to surface defects such as water seepage, cracks, and spalling during long-term service, seriously threatening driving safety and structural durability. Existing tunnel defect detection methods still lack multi-scale feature extraction and adaptability to embedded deployment. To address these issues, an improved tunnel-defect detection method named YOLOv11-WTseg was proposed. Specifically, a wavelet transform convolution block (WTConv) was introduced into the segmentation branch of the YOLOv11-seg architecture to enlarge the receptive field and enhance multi-scale feature representation. Meanwhile, the bounding box regression loss was replaced from complete intersection over union (CIoU) to Scylla intersection over union (SIoU) to optimize the regression process through an angle-based penalty mechanism. Based on a self-built tunnel defect dataset, comparative experiments were conducted on gutter damage, non-gutter slab damage, pavement cracks, water seepage, and vault spalling. Results show that the introduction of WTConv increases Box_mAP from 0.821 to 0.844, and Mask_mAP from 0.787 to 0.820. Further employing SIoU increases Box_mAP to 0.845 and Mask_mAP to 0.839. Compared to the baseline model, the improved model demonstrates superior detection and segmentation accuracy across various input scales. Furthermore, embedded deployment verification on the NVIDIA Jetson Xavier NX platform demonstrates that the improved model achieves both lightweight and real-time performance while maintaining a model size of only 5.4 M, achieving an inference latency of 117 ms. These results demonstrate that the YOLOv11-WTseg model can efficiently and accurately detect and segment typical tunnel defects and has promising engineering application prospects, providing technical support for tunnel structural health monitoring and maintenance decision-making.
LiuXuezeng, FangMaoliu, WuDexing, et al. Current status and development of rapid detection technology for tunnel structural defects[J]. Tunnel Construction, 2025, 45(4): 657-676.
NiuDaoan, KeZaitian, LiuWeizhen, et al. Research on the framework of high-speed railway infrastructure inspection and monitoring system[J]. China Railway, 2020(10): 9-17.
[7]
LongS H, YangT, WuY P, et al. GPR imagery based internal defect evaluation system for railroad tunnel lining using real-time instance segmentation[J]. IEEE Sensors Journal, 2024, 24(21): 35997-36010.
YangHao, GuoKaili, ZhangJunhui, et al. Quantitative identification of road interior voids based on YOLO11-L algorithm[J/OL]. (2025-08-18)[2025-09-20]. https://kns.cnki.net/kcms/detail/61.1313.U.20250818.1134.002.html.
FangLihan, ZhangQingwen, LiWeiguo, et al. A precise detection method for tunnel lining defects based on improved YOLOv5[J]. Science Technology and Engineering, 2025, 25(16): 6812-6820.
BaiZiyi, MaShuqi, YaoXiangchen, et al. The image segmentationmethod for tunnel leakage based on the improvement of U-Net network[J]. Science Technology and Engineering, 2025, 25(32): 13998-14008.
[18]
HeL, ZhouY, LiuL, et al. Research and application of YOLOv11-based object segmentation in intelligent recognition at construction sites[J]. Buildings, 2024, 14(12): 3777.
[19]
KhanamR, HussainM. YOLOv11: an overview of the key architectural enhancements[J]. arXiv Preprint, arXiv: 2410.17725, 2024.
[20]
WangA, ChenH, LiuL, et al. YOLOv10: real-time end-to-end object detection[J]. Advances in Neural Information Processing Systems, 2024, 37: 107984-108011.
[21]
FinderS E, AmoyalR, TreisterE, et al. Wavelet convolutions for large receptive fields[C]// European Conference on Computer Vision. Cham: Springer, 2024: 363-380.
[22]
GevorgyanZ. SIoU loss: more powerful learning for bounding box regression[J]. arXiv Preprint, arXiv: 2205.12740, 2022.
[23]
ZhengZ, WangP, LiuW, et al. Distance-IoU loss: faster and better learning for bounding box regression[C]// Proceedings of the AAAI Conference on Artificial Intelligence. Philadelphia: AAAI, 2020: 12993-13000.
[24]
BolyaD, ZhouC, XiaoF, et al. YOLACT: real-time instance segmentation[C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. New York: IEEE, 2019: 9157-9166.
[25]
SunP, ZhangR, JiangY, et al. Sparse R-CNN: end-to-end object detection with learnable proposals[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2021: 14454-14463.
[26]
RenS, HeK, GirshickR, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. Advances in Neural Information Processing Systems, 2015, 28: 91-99.
[27]
CaiZ, VasconcelosN. Cascade R-CNN: delving into high quality object detection[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2018: 6154-6162.
[28]
WangC Y, BochkovskiyA, LiaoH Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2023: 7464-7475.
[29]
SohanM, Sai RamT, Rami ReddyC V. A review on YOLOv8 and its advancements[C]// International Conference on Data Inte-lligence and Cognitive Informatics. Singapore: Springer, 2024: 529-545.